Computer-based applications generally contain textual data that is presented to a user to allow the user to interact with the applications. The textual data is written in a human language, usually the native language of the country of the developer or a common language, such as English. When the application is destined for use in other countries, typically the textual data is manually translated into the various local languages as part of a process commonly referred to as “localization.” Application metadata is used to store the structure and format of user forms, reports, queries, and other application components. This type of metadata may be stored in a repository. The human language user interface data is referred to as localizable metadata within the application metadata. The translation of the metadata may then take place.
However, translation and re-translation are time consuming, expensive, and require a long lead-time. Therefore, performing any actions that require a retranslation after a translation has been performed is highly inefficient. Even minor changes to the repository may include changes to translatable properties of an object, thereby requiring another expensive pass through the metadata to perform a retranslation.
One prior art solution freezes the entire repository or project after translation. This prevents any changes from being made, thereby preventing the need for retranslation. However, it prevents any changes to the repository/project near the end of a release. Therefore, this solution is not optimal.